Image analysis – Image enhancement or restoration – Intensity – brightness – contrast – or shading correction
Reexamination Certificate
1999-05-11
2003-03-18
Lee, Thomas D. (Department: 2724)
Image analysis
Image enhancement or restoration
Intensity, brightness, contrast, or shading correction
C382S254000, C358S406000, C358S504000
Reexamination Certificate
active
06535649
ABSTRACT:
BACKGROUND OF THE INVENTION
1. Field of the Invention
The present invention relates to a calibration method for color image devices, and in particular to a method of calibrating the color image devices dynamically in order to stabilize the output image.
2. Description of the Prior Art
Many different types of devices exist for sensing color images. These devices can be desktop or hand held scanners, copy machines, facsimile machines, or other similar devices.
FIG. 1
shows the interior structure of a transparent scanner
10
in which a light source
20
illuminates image
36
which is reflected by mirror
22
onto transparent lens
24
and then focused onto a photo detector
26
. The optical signal received by photo detector
26
is converted into a current signal and output to an analog-to-digital converter (ADC)
28
. ADC
28
will then transform the current signal to a digital signal and output the digital signal to an applied specific integrated circuit (ASIC)
30
. The data in ADC
30
is stored in memory
32
, and also output to host
40
through interface
34
. The data from ADC
28
can be handled by firmware, hardware or a host's software.
The main device within the scanner is photo detector
26
shown in FIG.
1
. Photo detector
26
is usually a charge coupled device (CCD) with a linear characteristic. In accordance with what is shown in
FIG. 2
, a photo detector
26
has N sensors, in which each sensor corresponds to a pixel whose width is the degree of optical resolution in units of dots per inch (dpi). Before starting scanning, a scanner will firstly scan a calibration target in order to get a reference of corresponding white or dark for obtaining the reference points conveniently. This procedure is called calibration.
FIG. 3
shows a calibration target
60
which is composed of a white calibration area
62
and a dark calibration area
64
. Their functions are defined as white reference and dark reference respectively.
FIG. 4
shows an output result of the calibration target after being scanned by a photo detector, in which T(
1
,
1
) represents the first pixel scanned by sensor S(
1
) and T(M, N) represents the Mth pixel scanned by sensor S(N).
An ideal situation would be for all Ts to be identical. Therefore, the conventional method fetches a single point only. However, noise within systems, changeable light source size, dust on the calibration target, attachment of fibers or 20 unknown particles, uneven lighting tube distribution, and system noise enlargement all affect interference during calibration. This causes the calibration result to diverge from the ideal value, and the scanned images to end up with streaking and blocks of not well-distributed phenomena residing on top of them, as shown in FIG.
5
. The vertical axis of
FIG. 5
is the brightness size, having “LEVEL” as its unit, wherein 0 is the darkest and 255 is the brightest. The horizontal axis represents the position scanned by the sensors.
FIG. 5
a
represents noise interference.
FIG. 5
b
represents lighting not well-distributed.
FIG. 5
c
represents having light gathering phenomena.
FIG. 5
d
represents having dust on top of the calibration target.
FIG. 5
e
represents an ideal situation where brightness versus coordinates is a constant.
FIG. 6
is an alternative expression of
FIG. 5
, wherein its vertical axis is “LEVEL” and its horizontal axis is “COUNT”. This expression defines the number of points counted relative to its relative brightness level. It is shown clearly that
FIG. 6
e
is an ideal situation, wherein the figure shown is a delta function.
A well-known prior art solution to the problem of divergence of the calibration result due to interference by dust is taking a simple average of multi-points, or taking an average by skipping adjoining sampling points. As shown in
FIG. 7
, photo detector
26
reads in the value of oblique-line portions from the calibration target and then takes an average. This method can evade a portion of interference-causing dust, but it is unable to deal with bigger dust or interference caused by different scanners, and also is unable to deal with signals coming from different sources of interference.
SUMMARY OF THE INVENTION
In accordance with the present invention, a method is provided for dynamic calibration that substantially solves all the drawbacks caused by conventional methods by eliminating or substantially reducing the effect caused by interfering factors in order to obtain good output images.
In one embodiment of the present invention, different sampling methods are used by different scanners in order to conform to the different scanners' characteristics. The sampling method can be fixed position multi-sampling, continuous position multi-sampling, discontinuous position multi-sampling, alternatively using fixed position multi-sampling and discontinuous position multi-sampling, or a multi-pixel averaging method. The sampling method is different in accordance with the actual scanner's characteristics.
A further embodiment of the present invention involves analyzing calibration information against different interfering factors. The calibrating information analyzing method can be a direct average method, weighting average method, fixed-range method, or standard deviation range method. The choice of method is decided by the distribution diagram of the multi-sampling.
Furthermore, another embodiment of the present invention uses the calibration method to decide system standards such as the calibration target standard, manufacturing process and environment standard, and a machine's maintenance time.
In accordance with the above embodiments, the present invention provides a dynamic calibration method. This method uses a photo detector to sample the first row of the calibration target and uses the result to obtain a standard deviation &sgr;
t
. A sensor S(N) within the photo detector multi-samples the calibration target and uses the result to obtain a standard deviation &sgr;
s
(N). Comparing those two standard deviations, if &sgr;
t
is larger then the direct average of the result of multi-sampling; and if &sgr;
t
is smaller then examining the distribution diagram of the multi-sampling and selecting a suitable calculating method with respect to different distribution diagrams.
REFERENCES:
patent: 5798847 (1998-08-01), Aerts
patent: 6198845 (2001-03-01), Tse et al.
Lee Thomas D.
UMAX Data Systems Inc.
LandOfFree
Dynamic calibration method does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Dynamic calibration method, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Dynamic calibration method will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3082539